Fort Meade, MD --
Service Acquisition Executive believes failing and fielding fast are key to building Air Force of the future
In early 2019, Airman interviewed Dr. Will Roper, the assistant secretary of the Air Force for Acquisition, Technology and Logistics.
As the Air Force’s Service Acquisition Executive, Dr. Roper is responsible for and oversees Air Force research, development and acquisition activities totaling an annual budget in excess of $40 billion for more than 465 acquisition programs.
In this position, Dr. Roper serves as the principal advisor to the secretary and chief of staff of the Air Force for research and development, test, production and modernization efforts within the Air Force.
Dr. Roper discussed how reforming the acquisition process is foundation to building the Air Force we need to maintain dominance in the battlespace today and tomorrow.
Airman Magazine: It feels like a tectonic shift is occurring in so many areas of the Air Force and acquisitions reform is the foundation that everything is being built on. Can you address why and how acquisitions reform is so integral to where the Air Force is going?
Dr. Roper: You think acquisition reform would be right next to watching wallpaper dry in excitement. A lot of people told me that I wouldn’t have as much fun in this job when I left OSD (Office of Secretary of Defense) doing a lot of rapid prototyping.
I think the myth’s been dispelled. Acquisition reform is exciting because we’re building amazing systems that no one else in the world gets to build. What we’re trying to do is just get back to our roots as being a rapid builder and fielder of things.
Speed does not sound like it ought to be the thing you focus on most in acquisition because we track everything in terms of dollars and cents. But the thing that jumps out when you’re working on something no one else has ever built is that you need to get out and start building as quickly as possible.
Because that’s when you start learning about the new thing you’re trying to accomplish. So, the emphasis on speed isn’t just about being fast and getting to the goal line, though that is really important – we have to field things faster than adversaries, it’s about getting information back into the programs to enrich our understanding, refine the direction we’re going and make better choices.
Traditional acquisition just takes a long time to do that. It’s not that that process is entirely bad. It’s good for extremely complicated weapons systems, but most things need a different path.
Airman Magazine: I don’t want to go through an entire primer on contract and acquisition speak here, but the difference between the waterfall method and what is being termed today, agile acquisitions, or DevOps, what are the differences and benefits of each?
Dr. Roper: It’s a good question because one thing it’s important to get your head around is that acquiring something that’s mainly hardware is very different than acquiring something that’s mainly software.
I think that’s intuitive as we look at the commercial world that software companies feel very different than companies that build a car or an airplane for you. We should see that reflected in our program. So let me give you the compare contrast between both of them.
Both can be rapid, both can be agile and we’re focused on doing both well because we have a lot of hardware in the Air Force, but increasingly software is where our lethal edge comes from.
So we’ve got to become world class and doing software is a big climb for us on the hardware side. There’s an old lesson the Air Force has had in its roots ever since it was created that prototyping is a great way to acquire something.
We like to talk about flying something before we buy it, but we don’t do that very often. We like to commit to buy something and then we’ll go make it and fly it and hope it works.
Well, prototyping is the step between where you know enough to go work with industry to tell them, “I want a plane. I wanted to fly this fast. I need it to do this mission.” But you don’t know how to turn your desire into 100 pages of specifications that everyone needs to be checked off.
You’re letting industry come back with what they can do. Allow yourself to learn as you go, and then after you test, you decide was the thing you flew good enough to buy or do you need to spiral again?
So we’re just rediscovering that prototyping is a lot better than spending three years thinking and analyzing and doing the analysis-paralysis approach before you actually get that first contact with industry and discover what they can actually make.
That’s a lesson that I think when we have programs come through and field faster. I think historians will look and say that is an old lesson that the Air Force has rediscovered.
Software’s a new lesson.
When I was trained, most of the people working in the Air Force were trying to think of software as a product. Software was a thing that was delivered years from now and you thought of it as a set number of lines of code that needed a lot of testing and certification before you deliver it.
Of course we’ve all seen, over the last 10 years, that software is now something that can change every week, every day. For some companies, hourly software changes are not unheard of and that’s because there are tools today that allow software to be developed in a distributed fashion.
The tools are auto checking the software, making sure that different coders code can work together and it doesn’t shoot each other in the foot. They’re able to synergize and that speeds up software delivery. Commercial developers are trying to push software out and get user feedback.
So, you know how your app on your phone updates? Those developers are often getting real-time feedback – are you using your app? Are you happy with it? Is your use duration commensurate with what they designed it for? Then they use that feedback to update.
The war-fighting analogy for that is obvious. We’d really like to be able to have software go out into the battlefield, get feedback from the users. What was day one of the war like? Change it, upgrade it and be different on day two. We have to adopt these commercial development tools that allow us to change software routinely, but change it safely.
And this is what we call DevOps. You’re developing, but you’re developing with the operator. The operator is your metric. That may sound like it’s wholly new, and I did say it’s a new lesson, but it’s a lesson where there’s an analogue in other parts of the Air Force – We don’t just acquire products, we acquire services.
IT (Information technology) is something we buy a lot of in the Air Force. How do you know if you’re getting good IT? Well, you’re never done with it? It’s not a product that delivers and finishes. You have metrics you look at are you getting the reliability? Are you getting the bandwidth? Are you happy with your experience? We use those to determine if we are getting good service or bad service?
Software development has its own metrics. Are you writing code quickly? Are there a lot of deficiencies? Are you able to retire those deficiencies quickly?
(Think of) software development not as a product that delivers you something, like an aircraft or a satellite. Think of it as a service where you have a pipeline of coders continually updating your system, making it better, refining things based on what users want.
You’re measuring their output with metrics that tell you, are they efficient? Are you getting good value for your money? As if they were giving you IT or a service like that.
Then we’ve got people in the Air Force that acquire services every day, billions of dollars of services. So the thing we’ve got a master is, we’ve got to take all of our good lessons from service acquisition and put those software unique twists in them so that we know how to talk software coding metrics with developers and know are we getting good value for our dollar?
Airman Magazine: With the warfighters being involved in constructing the weapon that they use in the battle space, could this possibly have a positive effect on retention?
Dr. Roper: I think ultimately warfighters are always the customer. It’s a shame that there are a lot of warfighters that have gotten accustomed to when they give feedback about a system, they’re giving feedback so that their successor or their successor’s successor can see the benefit.
The idea that an operator could ask for something and get it in weeks is mind blowing, but it shouldn’t be.
That should be the norm. A lot of times we focus on the coder. I don’t want the operator to be underplayed. That coder has no idea what to do unless the operator is able to say what the next capability spiral is.
The operator is substituting for the requirement stuff. The operator’s need becomes the requirement.
As the acquisition guy, it tells me we better have superior operators working with our coders or likely to be working on things that aren’t the highest priority to their commanders.
One of the things I’ve asked of all the Air Force MAJCOMs (major commands) is, when we’re doing these DevOps programs, we need to make sure the operators we’re working with can speak for the commander, because the thing we’re going to make is going to be in the image of the operator we’re working with.
Airman Magazine: We’ve talked a lot about doing things quickly and that seems very counterintuitive. I’ve heard you speak in the past that while it’s a priority to be a good custodian of money, the money may be better served if we focus on time. Could you describe how reducing the time to field a product or system can actually save money?
Dr. Roper: I talk about speed a lot – sustainable speed. It’s not the only thing in acquisition, but it’s the only thing you need to do to win. Speed is the only thing that’s going to ultimately matter.
The most efficient side is probably not going to be the winning side. The most efficient side is not going to be taking enough risk to continually move the envelope. Being the world’s second best Air Force probably has decreasing value in the future.
I feel passionate about speed for a variety of reasons. It is important to understand them. Some of them are strategic and some of them are very tactical, just related to doing good program management and good custody of taxpayer dollars being spent well.
The strategic side of speed is simply that we’re in a competition.
We’re leaving a period where we’ve been focusing solely on violent extremists and we’ve now got to compete with adversaries that can bring the same technological knowhow, the same military knowhow, and build systems that commensurate with the ones that we can build.
When you’re in a competition with a peer competitor like China or Russia, and you’re upgrading your systems but they’re upgrading at a faster pace, you might be better now and you might be a better five or 10 years from now, but the time constant between your upgrades is going to eventually give your competitor the advantage, just because they get more cycles.
But here’s the thing that’s counter intuitive about speed. When you focus on it as the primary objective, you tend to get smarter acquisitions too. If you think through it, it makes sense. You can’t be fast in an undisciplined way. If you’re a program manager and I tell you how to build a hypersonic weapon and go as fast as you can, there might be someone that thinks you’re going to chuck your cost estimates out the door. You’re not going to do any engineering analysis or risk analysis because you’re trying to save time.
But, when I tell you to deliver fast, you’re going to do absolutely the opposite. Those assessments, those analyses, they become a religious importance to you because … those things help you guide your program through the down slopes of risk management. You put a high priority and you must know them.
Those bureaucratic steps that are meant to just keep the paperwork boxes checked, that our defense acquisition systems so loves, those are the things that you absolutely cut out of your program with extreme prejudice. You can’t stand wasted time.
So the rigor goes up and the waste goes down, which tends to give you a more efficient path to running your program. It also forces you to think creatively.
So we’ve got programs now like the B-52 re-engine program, a program that came in as a traditional acquisition when I was early in this job. We were going to go buy it the DoD 5000 way, do 3,000 pages of studies and look at industry concepts before we really get down to working with industry.
Now that program is being bought the same way a commercial airline would choose its engine. And we’re going to have a digital (virtual) twin fly-off and do prototyping and compete industry competitors against each other. Have a lot of data in our hands before we pick the system. Well, which path do you think is smarter?
I look at the fast path and think, you’re going to have data from a digital twin. You’re going to have modeling. You would never have in a 5,000 series acquisition. So my money’s on the faster path.
Airman Magazine: And that reaches back to the concepts of failing forward embracing a simple scientific method – make your mistakes in the beginning so you know what not to do?
Dr. Roper: Absolutely. Seventy percent of a program’s budget is once we are, are sustaining it once we’re owning it in the field. So that should tell you that you need to get any failures, any flaws, any things that drive cost in your program – they need to be experienced early.
That’s why I’m a big believer in prototyping. It’s different than science and technology that’s kind of experimenting in a lab and it’s different than committing to a program of record that you’re going to keep many cases for the Air Force for decades.
Prototyping is a safe place to fail. The big litmus test for this going fast, and it’s still an experiment. We’ve just started all of this acceleration. We haven’t had those big blow ups yet. They’re going to happen. I’m not naive on this. We are going to have prototypes that don’t work.
The big question is going to be, does Air Force leadership; does Pentagon leadership, does congressional leadership look at that as a bad thing, or do they look at it and say, “thank God we didn’t make that a major defense acquisition program that we’re now stuck with those flaws for decades?”
Well, I can tell you the current Air Force leadership believes that finding those flaws in prototyping is a good thing. We’re going to be fierce defenders of those programs over on The Hill (Capitol Hill).
The thing we’ve got to explain, in plain English, is that by finding them there, we didn’t get committed to the big cost, which is when we move into production and sustainment. So find it early, fix it if you can, if not, terminate and move on.
Airman Magazine: What part does the Legislative Liaison Office play in educating staffers and representatives on Capitol Hill, who may be used to the Waterfall Method of acquisitions, as to the benefits of a more agile DevOps approach?
Dr. Roper: We work a lot with the Legislative Liaison Office because a lot of the questions that the members of Congress and staffers have are about acquisition programs. It’s a great partnership. Those people get us smart about what members of Congress are interested and in what programs and I think by the time people rotate out of their assignments. They’ve gotten pretty smart on Air Force programs.
I would say right now we’re really fortunate to have members of Congress and staffers who are pretty acquisition savvy, who are strong believers in keeping the dominance of the military and our Air Force. They have seen the acquisition system, because of all of its statutes and regulations continually fail to keep up with a world that’s changing rapidly.
The authorities that we’re using to speed things up have already been granted to us by Congress. And here’s a secret – they don’t actually give us a ton of truly new authority.
The Section 804 authority, if you boil it down, gives you two new authorities to do acquisition. One is a prototyping authority in one’s field. Prototyping allows you to tailor your acquisition plan, not worry about extreme documentation and learn by doing, which is separate from deciding to go buy that thing.
The fielding authority says if it’s something you can go get out in the field in less than five years, go buy it and get it fielded at as quickly as possible.
We’ve been using these two authorities to take programs that would have normally gone through traditional acquisition and speed them up by tailoring out all the steps that are unnecessary. Just think of it as a tailored suit fits better than a store-bought suit. It’s no less a suit, it just fits better to the needs of the person wearing it.
Here’s the secret, the authorities to tailor reside in traditional acquisition to.
There’s almost nothing that’s new in the 804 authorities. So why do our 804 programs look so different than traditional acquisition? If you step back from them it’s not an authority difference, it’s a culture difference.
We had the ability to do it, but it wasn’t clear that Congress would support us if we started doing things differently and by passing new legislation that says, “do things differently,” it encourages people to take the risk and take the step.
So I’m using the 804 for authority, not because it’s the only way to speed up acquisition, we had just as many tailored traditional programs. I don’t have to be something designated as a “special acquisition” to do things special in my program. I can be as special as I’ve got the daring instinct to be.
We’ve used 804 to encourage the culture change, to encourage program managers to step back from their programs and say, what am I doing that’s not necessary? Or what things could I be doing differently to speed up?
When someone says, “hey, you’re taking a risk,” I know that I’ve got the top cover from the Air Force and from Congress to take that risk if it’s smart.
We have such great talent in the Air Force, we have smart people managing programs that talent is being unchained and it’s thriving.
So if our experiment works and we ended up speeding up acquisition, I don’t think people are going to talk about the special authority that Congress granted us, even though we’re very happy about it. I don’t think they’re going to talk about the fact that we started doing a lot of prototyping. I think they’re going to talk about we put the reigns of programs back in program manager’s hands. And when we did that, because of the talent and training that we have in Air Force acquisition, those people took the reins and they ran. And I’ve seen that in programs across air and space and cyber. We’ve got a stacked deck across the Air Force. So why not bet on it.
Airman Magazine: It seems that all the planets have aligned here perfectly. You’ve got top cover from the chief of staff and the secretary and you’ve got people on the hill that want to play ball this way. Does it feel like a really huge point in Air Force history?
Dr. Roper: That’s why I enjoy coming to work each day. I thought long and hard about taking this job because I had a great job before at OSD, but I knew (Air Force) Chief (of Staff) Goldfein from the time he was the director of the joint staff.
I’ve been doing projects for him off and on for six years and have the greatest respect for him and I know he wants to see change in the Air Force. He sees the world changing and the Air Force, historically, has been the service that has taken whatever the new technology is, and we’ll find a way to weaponize and fight with it.
We’ve been the service that didn’t have a past; we had a future. The future was our legacy.
As I got to know Secretary (of the Air Force) Wilson, I was just extremely impressed by her sharing the chief’s (CSAF’s) vision, but having her own vision for increasing talent and training and streamlining acquisition, getting the Pentagon out of the day-to-day in programs and equipping the field to execute.
So when you’ve got leaders that you work for, where you’re aligned with their vision and talent in the workforce that is ready to do more than they’re currently equipped to do, who wouldn’t want to do that job?
Airman Magazine: Would you agree that DevOps dovetails nicely with the chief’s (CSAF’s) priority of pushing command-level decisions as far down the command chain as possible? Can you tell us a little bit about the different categories of acquisitions and what wings and squadrons have the ability to do now in acquisitions?
Dr. Roper: Sure. The theory is, push the decision as low as you can with acceptable risk. You don’t want to put people in a position to make decisions that they haven’t been trained or equipped to, but the way acquisition was before the reform, you had to become a colonel before you had a chance to make your first big mistake.
One, that’s way too slow and two, that’s a disservice to that senior materiel leader, that O-6, who didn’t get a chance to fail at smaller levels to become a better manager, a better risk taker, a better strategist in how they approach risk. So the idea is push, push the authorities down as low as we can.
We break acquisition into three categories.
In Category 1 are big programs. They’re the ones that you’ve heard of – the KC-46 tankers and large satellite programs. They still come up to me for approval but are few and far between.
That’s for major milestones and it’s not that I’m going to add anything of significance to them. The program managers are going to know so much more (about the programs) than I do. It’s really to help that person troubleshoot the political environment that a big program has to deal with, and sometimes, in the case of the period we’re in now, helping them understand new authorities or tools that they could bring to bear that maybe were part of the program when they inherited it.
So everyone needs to do agile software development. Why should a program have to relearn lessons that Kessel Run has already learned?
So that’s the kind of work I’m trying to do up at up at my level, is to make tools and new resources available so that when a program manager comes in and says, “here’s how I’m going to do my program,” I can say, “that’s awesome. Here’s a new tool. Now you can do it even faster!”
The Category 2 programs, ACAT-2, are smaller programs. Those are all delegated to our program executive officers. I don’t have a single one at my level. They don’t have to come to the Pentagon for any reviews; they can make decisions and run at the speed that they can go.
There are ACAT-3 level programs, there are hundreds of those in the Air Force, and those are delegated down even further to deputy PEOs and even program managers.
I hope in time we’ll even think about pushing decision making even lower.
So think about the change. We went from having one milestone decision authority in the Air Force, which would be the service acquisition exec, and now we’ve got 100.
We’ve increased our decision-making ability by orders of magnitude. What’s smart from a training perspective, is (not only) we’re getting people authorities earlier, responsibility earlier, which will make them better leaders as they move up the chain, but it actually increases our speed and agility.
We can make more decisions, which means we’ll be faster in getting things done.
Airman Magazine: Is there an effort being made to get new technologies out of the labs more quickly and into the hands of the warfighter?
Dr. Roper: There is. I think phase two of getting the reform done in Air Force acquisition starts by reminding ourselves that technology’s a big part of acquisition.
We are working very hard to try to up the transition rate out of our research lab and that’s going to be work that remains to be done at the beginning part of this year.
But, I tell you, my experience to make transition happened when I was in the office of the secretary of defense is that you need to be very strategic about what your high tech prototypes are.
There is a propensity, in most organizations, to peanut-butter resources too broadly so that you have a lot of smaller things, but they never really retire all the risks that are needed to create new programs out of them.
There’s some extreme discipline that’s needed to say, “I realize I have 50 needs in the Air Force, but I have to focus on five big things with my research portfolio if I’m going to have a valid attempt to get them into programs.”
Given that was my job when I worked for the SECDEF, it’s my hope to bring some of that to the Air Force. I’m going to have to partner with the research lab and with the Air Force Material Command commander and the MAJCOMs. There are a lot of people who have to be involved with that.
The other thing that I have to be honest about, as the acquisition exec, is a lot of innovations are happening in private industry now. That’s different than the Cold War, when the department drove a lot of the technologies that eventually were commercialized – satellites, internet, microprocessors; these were homegrown.
But now we see great innovation happening in startups and commercial industry and right now our acquisition system is too slow to work with them.
Heretofore, that’s just been something that we’ve said, “Oh, that’s a shame,” but I view it as a national security risk to not be able to work with this vibrant ecosystem of startups that are generating concepts that are literally changing the world.
The fact that most new technologies are highly inspired or driven by software has to break down the model that you need to be a large laboratory of 1000 people to make change. You can be a startup of 20 and 25 people and change the world. So we need to work with them. We need them to think about Air Force programs and Air Force problems. We need them to think of us as a partner and that’s something we’re really excited to do as we move into Pitch Day later this year.
Airman Magazine: So what kind of technologies are we talking about specifically that maybe we would be better off buying off the shelf or develop in partnership as opposed to creating them ourselves?
Dr. Roper: A lot of software-based solutions we should use commercial because they’re being developed quickly and they upgrade on their own.
We all have a big vision for how we use augmented reality and virtual reality. It can be things (for training) that might feel like (first-person combat video game) Call of Duty on the frontline of the battlefield, all the way back to the person who’s maintaining the system, being able to train them faster because they’ve got something that’s more immersive to get them up the learning curve as opposed to having to read and old school manual. I view that as something that we’re behind the times on.
Artificial intelligence – we’ve got to learn to leverage commercial technology. We’re not going to drive that technology. It’s going to be driven by large-scale commercial investment. So in acquisition, we’re going to have to learn how to harness and leverage something that we didn’t develop ourselves well.
The acquisition system that’s been bequeathed to us, born out of the Cold War, is used to the government creating its own technology that then gets operationalized. We were able to dictate the terms.
So this is a very different model. I don’t want to underplay the difficulty. We rarely will see a commercial technology be exactly what we want, but we’re going to have to learn to be okay with what it is and see the advantages that it gives and work around any potential weaknesses or vulnerabilities that may introduce. If we don’t, and we’re still doing everything internal, I’m willing to bet that’s going to be a losing strategy just given how quickly technologies are changing and private industry.
Airman Magazine: You mentioned Pitch Day. Some of these ideas come from very small businesses. Can you explain the concept behind Pitch Day?
Dr. Roper: We’re putting companies on contract the same day they pitch an idea using a government credit-card swipe. So rather than have to deal with all the accounting systems, if you have a PayPal account, you can work with the Air Force.
The thing that’s exciting about Pitch Day is that while the Air Force can buy products and services, we can’t really buy ideas, not for commercial startups. The reason is that a startup is small, 50 people or less, and it is really living paycheck to paycheck. They’re looking for investors.
Right now, it takes, on our best days, three or four months to get a company on contract; to get them money so they can work with us. That’s way too long for a startup that needs to have resource and investment today or this week at the latest.
So the mismatch in our contracting process means that we’ve got to work with big “started” (established) companies. They have to be big enough to already have their ideas and already be ready to sell them.
What we want to do is be able to influence a company that has a great idea, but doesn’t know who the customer is.
A lot of them are going to want their customer to be all of us as private citizens. They’re going to sell to the world. There’s no reason that the Air Force can’t be a partner in helping them get there.
So the idea with Pitch Day is to work with companies that are still formulating their ideas. They have a big problem they think they can solve or a product that they think has an application, but they don’t know what it is.
We want them to pitch it to us and have us help them find a mission, a national security mission they can solve.
Rather than say, “we love your idea, come see us in four months and we’ll be ready to work with you” and hope that they’re still in business, we have them leave today in partnership with the Air Force, ready to work with us on maturing (that idea).
We’ve got to make sure we do not tie that company so tightly within the defense contractors sphere that they can’t break loose and make it to be a viable commercial company.
Whatever the next generation of Facebooks or Googles or Amazons are, we want them to have a relationship with us before they’re that big. We want them to be savvy about our problems and know us, not because we want them to be another defense prime. We want them to be a national security savvy company.
Our hope is by being an excellent partner, an excellent incubator of ideas, that more companies will think about orbiting the Air Force before they reach escape velocity and go do big things in the commercial world.
Airman Magazine: So when and where is Pitch Day?
Dr. Roper: So we’ll do it the 6th and 7th of March in New York City and the idea is to make it just like a venture capitalist Pitch Day.
We’ll put problems out, they’re out online right now, and we have some specific things that we are interested in them solving, communication and battle management software needs.
We’re aware that battle management is not a term used by small companies in The Valley (Silicon Valley) and the bay area. So we’re also just open to cool ideas.
They’ll submit. We’ll look at them and invite a subset of companies to come to New York. If we’re impressed, they’ll leave with funding that day.
Airman Magazine: Speaking of agile acquisition, are there any specific success stories you would like to mention?
Dr. Roper: I think the success story that I’m so happy with, is that no one particular type of program appears to take the lion’s share of what agile gives.
You might expect a hyper-sonic program to benefit from going agile and rapid. Over five years of acceleration between where the traditional program was versus the rapid program. We’re going to try to hit the first test of a long-range hypersonic weapon less than two years from today.
That’s amazing. That takes it out of the future. It says, no, this is a “now” capability, but it’s “now” because we’re going to go out, test aggressively, be okay with failing, and if we fail and learn something from it and are able to make improvements, that’s not a failure. That success, that’s part of the learning process.
Most people do not expect that software programs would fall into the same bucket, but our cyber programs, like Unified Platform, in which we’re going to provide a consolidated cyber tool for all the joint war fighters, were originally going to deliver something four or five years from now.
They’re going to push out their first product in March or April of this year and it’s going to start a pipeline of deliver, deliver, deliver, on a quarterly basis. So rather than deliver the elephant at the end, we’re going to deliver a piece of it at a time, but using commercial development tools that make sure all those hunks of elephant truly add up to one elephant.
Everyone that is in the cyber world ought to be extremely excited because I am seeing our cyber program managers fully adopt agile DevOps for software and the warfighter. The benefit to the war fighter is that rather than wait, they start being able to get product now.
The great thing about the DevOps is the next delivery is what the operator says they need the most next.
Now this isn’t a statistic that crosses the whole Air Force, but for some of the software programs that we have, when I first joined, over 40 percent was not being used by the operator. That says if you’re not in close sync with the person using it and you’re going off doing traditional development, you may come back three or four years later and say, “here’s my delivery,” but the operator’s needs have changed. The mission has changed. The threat has changed.
There’s a reason why software in the commercial world doesn’t do that anymore. They do very short developments.
Even in our sustainment world, which is not part of the Air Force that we typically talk about, sustainment is 70 percent of what we do. Many of our programs are not developing anymore. They’re not designing anymore. They’re sustaining and ready to go fight a war today.
But there’re so many commercial technologies that can make these programs faster and cheaper and smarter. We’re starting to use predictive maintenance, have AI (artificial intelligence) algorithms that tell you when you need to replace a part before it breaks so that you save all of that downtime that is required when a system breaks.
We’re starting to 3D print parts ourselves. We don’t have to go out to suppliers if we can make it ourselves.
So even the side of the Air Force that rarely gets the limelight is getting a lot of limelight for me because these technologies are completely changing the game. And that’s the fun of doing this job. There’s so much talent here. There are so many technologies that are ready to be applied. The only thing I have to do is just be kind of “Match.mil,” connect people with tools that they don’t have and watch the magic happen.
Airman Magazine: Even when you are taking delivery of a new system, you’ve got to sustain the legacy systems that are still carrying out the mission. What are the challenges with balancing the two?
Dr. Roper: It’s a full-time business. You’ve got to be able to fight the war today. The most efficient way to maintain the plane will never be the way you do it, because your real objective is having a certain number of aircraft ready to fly.
The thing you see, almost immediately, when you come into the Air Force, having done development or production for most of your career and now there are a lot of sustainment programs, is you realize there’s no one in the Air Force who’s responsible for improving the technology in the entire sustainment enterprise. We’ve got something that’s responsible for doing that in development, that’s called our research laboratory and it’s responsible for making new technologies that every program gets to benefit from. But that’s for new programs, there’s no program, there’s no office, there’s no lab that does that for depots where we repair plans.
One of the things that was really clear that we needed to create a new program executive officer who’s just responsible for making sure that our depots, our maintainers, have the best technology available.
Some of the things that we’ve done, our transition and predictive maintenance, which uses artificial intelligence, we’ve transitioned 3D printers that can make parts and cold spray repair capability that can replace metal that’s been worn off.
These are things that we see in private industry. It just wasn’t anyone’s job to make sure that the Air Force had (those technologies) getting certified for use.
We’ve seen great, great returns on investment, even in the first few months of this office, because all of these technologies are shovel ready.
Airman Magazine: How are evolving acquisitions processes benefitting the Air Force in the space domain?
Dr. Roper: I think space is seeing some of the most creative and the most agile acquisitions that we have. It’s a shame that a lot of our space acquisitions end up being classified.
As we’re starting to think of space as a warfighting domain, we’re bringing in the same thought process we do for all the other domains – that perfect can easily become the enemy of good. We have to be able to field systems quickly. We have to give operators capability today, which means reaching for that 100th percentile is not a good idea if 95 can be done now. That pressure relief valve from 100 percent to 95 is where most of the speed of acquisitions comes from.
A traditional acquisition that has to hit the 100-percent bar is likely to have to take more time than you expect. It makes sense that it would, because you’re asking for something that’s likely never been built before, that’s pushing an envelope or a barrier. Why would you expect that you wouldn’t discover something’s harder than you originally thought?
Well, if you can’t drop the acceptable performance, then you’re going to keep putting money and time until you hit that bar, which is why programs often stretch out to be decades.
Now that we have to get ready to deal with contingencies in space, now we hit that hard thing when I tell the commander of Air Force Space Command, who’s been a great partner, “I’m sorry, general. I can’t give you 100 percent of what you wanted, but the industry base can give you 95 percent now. Do you want to field now or do you want to wait?”
It’s field now. Every time.
That is opening the door for creative strategies across the board, so I’m very proud of our space acquires, but I’ll tell you, I’m very proud of our space operators as well, because almost all of our space software development is DevOps and we have great operators who are excited to get to get software into their hands now, so big kudos to them.
Airman Magazine: On the testing, training and readiness front, there’s been a $235 million down payment on upgrading the infrastructure of ranges. Where does that priority stand in the acquisition process?
Dr. Roper: Ranges are critical for us because so much of what we do has testing involved and much of the testing that we need to do, especially on complicated aircraft and radars and things of that nature, you have to go do live testing.
It really is time to look at our ranges and modernize (them) so that we can make sure that the performance that we think we’re going to have in the real world, is really real.
Modeling and simulation is very valuable, but it only goes so far. It’s overdue in my view, but I’m glad that we’re doing it now.
The thing I worry about, as we create more and more systems, is just having enough time to test it all. The ranges are getting pretty crowded, but I underscore how important they are for our business.
Airman Magazine: We talked about failing fast and failing forward. That seems to be the chief of staff’s priority for individuals, too. So, I just wanted to ask you, was there a moment in your life were a failure actually propelled you to further success?
Dr. Roper: If you wanted me to go through all the failures, this will be a long interview. I’ll share a story that is really motivating for me in this job.
Before I joined the Air Force, when I was a new government employee working for Secretary Carter, I was given a little bit of money early on. It was just enough to place a bet on one system.
So I ended up taking a Navy interceptor called Standard Missile 6 (SM-6) and reprogramming it to do ship strike. So giving it a new mission it wasn’t designed for.
It’s about $10 million to do it. We went out and did a test and it ended up that the SM-6 was able to do the mission. It was a great success and everyone was very happy and it transitioned and is operational today.
I was very mindful when that test happened, that whether or not the missile had been able to do that job, had nothing to do with me or anyone else. It either could or it couldn’t.
So the fact that it ended up succeeding, it’s just luck. The fact that ended up getting additional funding and people was just luck because the system was able to do it. The thing I really should have been evaluated on was the risk itself. There was a chance to give us the Navy a completely different capability for a very low dollar investment.
The risk itself was a good choice, but the result, that I ultimately get judged on, was luck. I shouldn’t get any credit for that.
What I see here in the Air Force is that we often talk with people who are risk takers, who got lucky. You get people, like me, who love to talk about taking risks because we benefited from it. I realized there’s hypocrisy in that.
So I’m trying to be very mindful, very mindful, with Air Force program managers and people taking risk that they get their evaluation and validation for me at the point that they take the risk.
If they made a smart choice (on taking a risk), you’re a success, you’re a winner, you’re a rock star. Whatever happens because of it, that’s a roll of the dice. That’s just the outcome.
Why are the people who succeed the only people we cite when we talk about risk taking is a virtue?
If a program manager makes a $5 million investment to save $500 million, and it doesn’t pay out, isn’t that still a smart investment? If someone came in and gave you odds like that in Vegas, you’d be there tomorrow.
I realize I’m the beneficiary of a double standard that we like to talk about risk taking. We only reward the people for whom that risk works out. And that’s not the way that we should do it. A good risk is a good risk regardless of the outcome.
If you went back historically, I think most of the goals set would be about saving money.
So if you’re an engineer, how are you saving money? If you’re a contracting officer, how are you? Not everyone’s job traces to that. It doesn’t happen every day.
So one of the reasons I focused on time so much is that we contribute to the time and rapidity of programs every single day. Every person who is part of a program can do their job faster and be more agile.
So coming into the Air Force, I laid out a challenge to remove 100 years out of our acquisition portfolio. If you’re the program manager and get your acquisition strategy gets done a year early, you just contributed a year.
If you are the officer that gets that staffing package up to me a month earlier, you just added a month.
Every person is part of time and our lives are so driven by the time I think time is something that we understand intuitively. It has personal importance. It has importance to our war fighter. It has importance to the Air Force.
I’m going to be very focused on most of our challenges and goals being around time because I expect every person to look at their job and say, “I’m going to buy into this big vision the Air Force has and here is how I’m going to make my contribution.”
I’m seeing people that had been waiting for the stars to align, and I think they have with authorities and Congress and Air Force leadership saying, “we’re going to do this.” I see people that are taking the reins.
My experience has been the rock stars will take the reins and go. Then there will be a large pack that says, “I just saw what they did. I’m going to do it too.” I’ve seen the rock stars move out and now I’m starting to see the followers have impact.